11 Hits in 1.3 sec

Porting WarpX to GPU-accelerated platforms [article]

A. Myers, A. Almgren, L. D. Amorim, J. Bell, L. Fedeli, L. Ge, K. Gott, D. P. Grote, M. Hogan, A. Huebl, R. Jambunathan, R. Lehe (+11 others)
2021 arXiv   pre-print
We describe the strategy followed to allow WarpX to use the GPU-accelerated nodes on OLCF's Summit supercomputer, a strategy we believe will extend to the upcoming machines Frontier and Aurora.  ...  WarpX is a general purpose electromagnetic particle-in-cell code that was originally designed to run on many-core CPU architectures.  ...  Conclusion We have summarized the approach taken to porting WarpX, which was originally designed for many-core CPU architectures, to take advantage of GPU-accelerated nodes.  ... 
arXiv:2101.12149v2 fatcat:bldjocjryndnrgzovjvnegtbfu

Energy Applications Challenges (SIAM CSE21) [article]

Thomas Evans
map to SIMT architectures.Multiple programming models and approaches are being used to achieve performance portability across a range of GPU-based architectures.  ...  These areas include wind power, combustion, nuclear energy, carbon capture, fusion energy,and plasma accelerators.  ...  levels of refinement • Embedded boundary treatments between grid solve and particles • DEM algorithmic efficiency on GPU WarpX Challenge Problem and Codes High-fidelity simulation of whole-  ... 
doi:10.6084/m9.figshare.14125667.v2 fatcat:gvh4wndrrbh7vcgk7ily24fmdi

Energy Applications Challenges (SIAM CSE21) [article]

Thomas Evans
map to SIMT architectures.Multiple programming models and approaches are being used to achieve performance portability across a range of GPU-based architectures.  ...  These areas include wind power, combustion, nuclear energy, carbon capture, fusion energy,and plasma accelerators.  ...  Challenge Problem and Codes Modeling of a chain of tens of plasma acceleration stages resulting from laser and charged-particle driven wakefields • Base: 5 accelerator stages • Stretch: Code  ... 
doi:10.6084/m9.figshare.14125667.v3 fatcat:jiazayqlsjgbpaxe7hiolhukrq

HiPACE++: a portable, 3D quasi-static Particle-in-Cell code [article]

Severin Diederichs, Carlo Benedetti, Axel Huebl, Rémi Lehe, Andrew Myers, Alexander Sinn, Jean-Luc Vay, Weiqun Zhang, Maxence Thévenet
2021 arXiv   pre-print
HiPACE++ is a modular, open-source code enabling efficient modeling of plasma accelerators from laptops to state-of-the-art supercomputers.  ...  The longitudinal parallelization is done through temporal domain decomposition, enabling near-optimal strong scaling from 1 to 512 GPUs.  ...  (b) Lineout of the accelerating field from WarpX, HiPACE, and HiPACE++. The inset shows a zoom on the witness beam region, where flattening of the accelerating field due to beam loading is visible.  ... 
arXiv:2109.10277v1 fatcat:t6rmhd7ypbb2fdpb77sfrkrz4a

A New Kind of Data Centric Performance Portability Challenge Item [article]

Tim Germann
Some aspects of the very heterogeneous architectures that have recently begun to emerge, such as tensor cores, specifically target use cases such as deep learning.  ...  In that case, frameworks built upon abstractions that have emerged allow for low-level architecture-specific optimizations to be implemented and enable performance-portable codes.  ...  the XGC legacy PIC code CPU GPU CPU GPU 2019 2020 RK Step 1 RK Step 2 • More kernels were converted to C++ in 2020 and ported to GPUs: • Electron push: converted to C++/Cabana/Kokkos  ... 
doi:10.6084/m9.figshare.14125964.v3 fatcat:ppdp4wlqarhpdkdux3dpnwksle

sputniPIC: an Implicit Particle-in-Cell Code for Multi-GPU Systems [article]

Steven W. D. Chien, Jonas Nylund, Gabriel Bengtsson, Ivy B. Peng, Artur Podobas, Stefano Markidis
2020 arXiv   pre-print
PIC codes require new algorithm design and implementation for exploiting such accelerated platforms.  ...  We test sputniPIC on three multi-GPU platforms and report a 200-800x performance improvement with respect to the sputniPIC CPU OpenMP version performance.  ...  Our current task is to port the field solver or part of it (the residual calculation at each iteration) to GPU.  ... 
arXiv:2008.04397v1 fatcat:ay3g2c7rynhmxbu4uod56p2j5y

Snowmass21 Accelerator Modeling Community White Paper [article]

S. Biedron, L. Brouwer, D.L. Bruhwiler, N. M. Cook, A. L. Edelen, D. Filippetto, C.-K. Huang, A. Huebl, N. Kuklev, R. Lehe, S. Lund, C. Messe (+13 others)
2022 arXiv   pre-print
individual accelerator elements, to the realization of virtual twins that replicate all the complexity to model a particle accelerator complex as accurately as possible.  ...  After a summary of relevant comments and recommendations from various reports over the last ten years, this paper examines the modeling needs in accelerator physics, from the modeling of single beams and  ...  Some accelerator software have been or are being ported to GPUs and are being ready to run efficiently on upcoming Exascale supercomputers [234, 235] .  ... 
arXiv:2203.08335v2 fatcat:snrivaotkzdvznsjayvenfpg2i

Preparing Nuclear Astrophysics for Exascale [article]

Max P. Katz, Ann Almgren, Maria Barrios Sazo, Kiran Eiden, Kevin Gott, Alice Harpole, Jean M. Sexton, Don E. Willcox, Weiqun Zhang, Michael Zingale
2020 arXiv   pre-print
In this paper we describe the changes that have been made to these codes to transform them from standard MPI + OpenMP codes targeted at petascale CPU-based systems into a form compatible with the pre-exascale  ...  Astrophysical explosions such as supernovae are fascinating events that require sophisticated algorithms and substantial computational power to model.  ...  PORTING TO GPUS Our early attempts at preparation of the AMReX astrophysics codes for heterogeneous architectures focused on programming models that were minimally invasive.  ... 
arXiv:2007.05218v1 fatcat:x32sereeenf5niid65px6i5cci

Exascale applications: skin in the game

Francis Alexander, Ann Almgren, John Bell, Amitava Bhattacharjee, Jacqueline Chen, Phil Colella, David Daniel, Jack DeSlippe, Lori Diachin, Erik Draeger, Anshu Dubey, Thom Dunning (+39 others)
2020 Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences  
As noted in Wikipedia, skin in the game refers to having 'incurred risk by being involved in achieving a goal', where 'skin is a synecdoche for the person involved, and game is the metaphor for actions  ...  with the skin being exascale applications and the game being delivering comprehensive science-based computational applications that effectively exploit exascale high-performance computing technologies to  ...  The authors would like to extend a special thanks to the many computer and computational science researchers (hundreds of them) who have committed their time, talents, experience and passion to the ECP  ... 
doi:10.1098/rsta.2019.0056 pmid:31955678 fatcat:6gijc65k2fa5lirhdkiudbie5q

PIConGPU: Predictive Simulations of Laser-Particle Accelerators with Manycore Hardware

Axel Huebl, Michael Dr. Bussmann, Thomas Dr. Kluge, Ulrich Prof. Dr. Schramm, Thomas E. Prof. Dr. Cowan, Paul Prof. Dr. Gibbon, Burkhardt Prof. Dr. Kämpfer, Stefan PD Dr. Grafström
2019 Zenodo  
In this study, 3D modeling with the GPU supercomputer Titan enabled the identification of pre-expansion to near-critical target conditions, which uncovers a regime of volumetric laser-electron interaction  ...  Numerical studies are performed on two novel targets for laser-proton acceleration with near-critical and mass-limited properties.  ...  Measurements show the central PIConGPU routines after three weeks of porting to Alpaka for single and double precision.  ... 
doi:10.5281/zenodo.3266819 fatcat:b73a3zvxvjecjplcsks7gf52cu

Using DOE Math Libraries: Introducing the xSDK [article]

Lois Curfman McInnes, Jim Willenbring, Ulrike Meier Yang, All XSDK Developers
2018 Figshare  
describes the development of the xSDK, while at the same time explain the collaboration among math library teams toward two strategic objectives: building community and building sustainability—useful to  ...  We will discuss xSDK community policies that help to improve code quality and compatibility across independently developed packages, while also determining criteria for member packages' inclusion in the  ...  release - • Have a Spack package • Port to target platforms • Provide user support • Obtaining the latest release:  ... 
doi:10.6084/m9.figshare.6015977.v1 fatcat:4fn4sy5dbbbkzhzym57dtecu3m